Double concen a ion explaining he ou s anding inc ease in Spanish
c op p oduc ion
Miguel Ma ín-Re o illo (Ma ín-Re o illo, M)1, Ana Se ano (Se ano, A)2 and Ignacio Cazca o (Cazca o, I)3
1 Uni e sidad de Alcalá. Facul y o Economics, Business and Tou ism, Dep . o Economics, Plaza de la Vic o ia 2, 28802 Alcalá de Hena es, Spain.
2 Uni e sidad de Za agoza. Ag i ood Ins i u e o A agon (IA2). Facul y o Economics and Business S udies, Dep . o Economic Analysis. G an Vía 2,
50005 Za agoza, Spain. 3 ARAID (A agonese Agency o Resea ch and De elopmen ). Ag i ood Ins i u e o A agon (IA2). Facul y o Economics and
Business S udies, Dep . o Economic Analysis. G an Vía 2, 50005 Za agoza, Spain.
Abs ac
Aim o s udy: To e alua e he changes in Spanish ag icul u al p oduc ion since 1950s in a con ex o in ense ans o ma ions in e ms o
he egional and c op composi ion.
A ea o s udy: Spanish p o inces du ing he second hal o he wen ie h cen u y.
Ma e ial and me hods: We use index decomposi ion analysis o e alua e he changes in he alue and olume o c op p oduc ion, as well
as he ole o p oduc composi ion and he egional dis ibu ion o p oduc ion.
Main esul s: Spanish ag icul u e ha e ocused on ce ain egions in he sou h o in he eas o Spain. Some p oduc s like ege ables o
ui s ha e a posi i e p ices and composi ion e ec s, encou aging he p oduc ion in hese p o inces.
Resea ch highligh s: We ound a ‘double concen a ion’: Spanish ag icul u e has inc easingly ended o p oduce high alue-added p o-
duc s, such as ege ables, ui and oli e oil. On he o he hand, c op p oduc ion is concen a ed in he sou he n and eas e n p o inces o
Spain.
Addi ional key wo ds: Spain; ag icul u al sec o ; ag icul u al de elopmen ; index decomposi ion; egional analysis
Abb e ia ions used: CAP (Common Ag icul u al Policy); CE (Composi ion E ec ); EEC (Eu opean Economic Communi y); IDA (In-
dex Decomposi ion Analysis); LMDI (Loga i hmic Mean Di isia Index); NRA (Nominal Ra e o Assis an ); PE (P ice E ec ); QE (Quan i y
E ec ); RE (Sha e E ec ); SE (Scale E ec )
Au ho s’ con ibu ions: These au ho s con ibu ed equally o his wo k. All au ho s ead and app o ed he inal manusc ip .
Ci a ion: Ma ín-Re o illo, M; Se ano, A; Cazca o, I (2020). Double concen a ion explaining he ou s anding inc ease in Spanish
c op p oduc ion. Spanish Jou nal o Ag icul u al Resea ch, Volume 18, Issue 3, e0107. h ps://doi.o g/10.5424/sja /2020183-15760
Supplemen a y ma e ial: (Table S1 and Figu e S1) accompanies he pape on SJAR’s websi e
Recei ed: 19 Sep 2019. Accep ed: 02 Jul 2020.
Copy igh © 2020 INIA. This is an open access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion 4.0 In e na-
ional (CC-by 4.0) License.
Compe ing in e es s: The au ho s ha e decla ed ha no compe ing in e es s exis .
Co espondence should be add essed o Miguel Ma ín-Re o illo: [email p o ec ed]
Spanish Jou nal o Ag icul u al Resea ch
18 (3), e0107, 12 pages (2020)
eISSN: 2171-9292
h ps://doi.o g/10.5424/sja /2020183-15760
Ins i u o Nacional de In es igación y Tecnología Ag a ia y Alimen a ia (INIA)
OPEN ACCESS
RESEARCH ARTICLE
Funding agencies/ins i u ions P ojec / G an
Ramón A eces Founda ion CISP15A3198
In oduc ion
Ag icul u al p oduc ion and p oduc i i y in Eu ope
and hei d i e s ha e been among he main hemes o
ag icul u al economics (Hayami & Ru an, 1985; G igg,
1992; O’B ien & P ados, 1992; Fede ico, 2005; Lains &
Pinilla, 2009; Als on & Pa dey, 2014; Gollin e al., 2014).
They de eloped s ongly du ing he second globaliza ion
( he second hal o he wen ie h cen u y), owing o majo
ans o ma ions in he sec o . These included echnical in-
no a ions ha inc eased dependence on he non-ag icul-
u al sec o s o he economy, g owing ma ke in eg a ion
o ag icul u al p oduc s and a educ ion in he sha e o
he labou and land ac o s o p oduc ion (while capi al
inc eased i s sha e).
The case o Spain is s iking o , among o he easons,
i s s eady inc ease in ag icul u al p oduc ion and p oduc-
i i y, especially du ing he second hal o he wen ie-
h cen u y (Cla e al., 2016). Such con inuous g ow h
was no seen in o he Eu opean coun ies, and since he
1980s ag icul u al p oduc ion has s agna ed (Ma ín-Re-
o illo & Pinilla, 2015). This g ow h was no sp ead
2Miguel Ma ín-Re o illo, Ana Se ano and Ignacio Cazca o
Spanish Jou nal o Ag icul u al Resea ch Sep embe 2020 • Volume 18 • Issue 3 • e0107
e enly ac oss he Spanish egions, and i also a ied
by p oduc g oup.
On a massi e scale, Spain in oduced echnical inno-
a ions om he non-ag icul u al sec o s o he economy,
such as ag icul u al machine y, chemical e ilize s and
hyb idiza ion and selec ion o seeds (G igg, 1992; Abad
& Na edo, 1997; Hassine & Kandil, 2009). Mo eo e , he
de elopmen o i iga ion boos ed ag icul u al p oduc ion
hanks o he g ea ad ances in wa e in as uc u e.
Spain’s ag icul u al mechaniza ion was as e han
ha o o he Eu opean coun ies, despi e adminis a-
i e obs acles o he widesp ead in oduc ion o ac-
o s in he 1950s (Cla , 2009).1 Thus, Spain con e ged
wi h he Eu opean le el o mechaniza ion (Ma ín &
Pinilla, 2015).
Second, chemical p oduc s such as e ilize s and pes i-
cides we e ano he impo an non-ag icul u al inpu . While
he in oduc ion o hese chemicals was signi ican h ou-
ghou Eu ope du ing he wen ie h cen u y (and especially
a e he Second Wo ld Wa ), in Spain i was done on a
huge scale, so ha once again he coun y con e ged on
Eu opean le els.
Thi d, as men ioned abo e, biological inno a ions did
much o b ing abou an imp o emen in ag icul u al p o-
duc ion and p oduc i i y. These ad ances, which began in
he inal decades o he nine een h cen u y (Olms ead &
Rhode, 2008), in ensi ied in he 1940s and 1950s, ocusing
on he hyb idiza ion and gene ic selec ion o seeds. The
mos de eloped coun ies all bene i ed om hese inno a-
ions, bu because o hei pa icula clima e condi ions, he
yield inc ease was lowe in he Medi e anean coun ies
han in Wes e n Eu ope (Pujol, 2011; Ha wood, 2018).
Finally, he expansion o i iga ion was also ele an in
a Eu opean Medi e anean coun y like Spain. On he one
hand, he echnical inno a ions ela ed o i iga ion in as-
uc u e allowed wa e o be supplied o a id egions: while
he a ea o land i iga ed mo e han doubled on he Eu o-
pean con inen gene ally, in Spain i almos ipled in he
second hal o he wen ie h cen u y. This expansion was
especially in ensi e du ing he 1950s, 1960s and 1970s,
when he F anco dic a o ship p omo ed he cons uc ion o
dams as a pilla o i s ag a ian policy. I iga ed p oduc ion
as a sha e o o al ag icul u al p oduc ion inc eased om
42.3% o 65.5% be ween 1955 and 2006 (Pinilla, 2008;
Cazca o e al., 2015a; Ma ín-Re o illo & Pinilla, 2015).
On he o he hand, he use o plas ic g eenhouses, sp inkle
1 Despi e he highe inc ease in ag icul u al mechaniza ion in Spain, he le el o mechaniza ion was lowe han he Wes e n Eu opean coun ies in he decades o he 1940 and 1950
(Ma ín-Re o illo & Pinilla, 2015).
2 The Nominal Ra e o Assis ance (NRA) ies o app oxima e o he pe cen age by which go e nmen policies ha e aised g oss e u ns o a me s abo e wha hey would ha e
been wi hou he go e nmen ’s in e en ion (o educed hem, i NRA < 0). In he empi ical wo k, he NRA is exp essed as a pe cen age o he undis o ed p ice (Ande son &
Valenzuela, 2008).
3 The second pe iod o analysis includes he en ance o Spain in he ECC, wi h he CAP in place. This in ol ed policies o suppo in p ices (subsidies) o some p oduc s such as
ce eals un il a ound 1991. F om 1992 he e was a eo ien a ion owa ds di ec aids o main ain a me ’s ag a ian income (coupled wi h p oduc ion un il 2003 and hen decoupled).
This implies some he e ogenei ies in he p ocesses o p ices c ea ion, in he same ashion ha di e en d i e s (ins i u ional, changes in openness’s and demands om o eign ma -
ke s, li es ock demand, e c.) ha e played a ole in p ices o ma ion. The de la ion used in his a icle akes in o accoun he gene al in la ion o all he c op p oduc s in he sample,
and hence i is mo e a e e ence which allows o dis inguish a di e en ial e olu ion o ela i e p ices among c ops, han a me hod o cap u e and disen angle hese d i e s.
and d ip i iga ion and sandy soils we e among he mos
impo an echnical inno a ions ha allowed an imp o e-
men in wa e p oduc i i y (López and Losada, 1999; Caz-
ca o e al., 2019).
Apa om hese echnical de elopmen s, he new
ins i u ional amewo k was also key o he expansion o
Spanish ag icul u al p oduc ion. Spanish ag icul u e ook
a s ep owa d ein eg a ion wi h he S abiliza ion and Libe-
aliza ion Plan o 1959, bu i was he coun y’s 1986 en y
o he Eu opean Economic Communi y (EEC hence o h)
ha p o ided he main s imulus. In he pe iod om 1960
o 1986, Spanish ag i- ood expo s g ew a an annual a e
o 5.8%, while impo s ose by 4.7%. This expo g ow h
coincides wi h a sligh inc ease in he openness a io du ing
he 1970s, and wi h a s ong inc ease om 1985 o he i s
yea s o he wen y- i s cen u y (Cla e al., 2015).
Spain’s en y in o he EEC no only eased comme cial
exchanges, bu also modi ied he le el o ag icul u al p o-
ec ionism. In e en ion in he ag icul u al sec o was e-
la i ely low un il Spain became a membe o he EEC. In
ac , he Nominal Ra e o Assis ance (NRA) was e y low
du ing he 1950s, 1960s and ea ly 1970s, compa ed wi h
o he Wes e n Eu opean coun ies.2 Basically, he common
ag icul u al policy (CAP hence o h) gua an eed high p i-
ces (gene ally highe han in e na ional p ices), which en-
cou aged ag icul u al p o ec ion om he ea ly days o he
EEC igh up un il he MacSha y e o m o 1992. Spain’s
low le el o ag icul u al p o ec ion inc eased om 1986,
when he coun y joined he EEC. Thus, he NRA g ew
om app oxima ely 10%, as a pe cen age o he undis o -
ed p ice in he i s hal o he 1980s, o a maximum o
70% by he end o ha decade (Ande son & Valenzuela,
2008).3
In sum, he lack o ag icul u al suppo and he dic a o-
ial con ex un il he la e 1970s bo h gene a ed an ag icul-
u al sys em based on he p oduc i is model. This model
ocused on inc eases in ag icul u al p oduc ion, he expan-
sion o i iga ion, he la ge-scale adop ion o non-ag icul-
u al echnical inno a ions, and modi ica ion o he egio-
nal and c op pa e ns o ag icul u al p oduc ion. All his,
combined wi h a high numbe o hou s o sunshine and
be e a ailabili y o wa e esou ces inc eased he sec-
o ’s po en ial in he ollowing decades. Since he 1980s,
wi h accession o he EEC and implemen a ion o he CAP,
his inc ease in p oduc ion has been linked o in e na ional
ma ke s – especially in he case o hose expo p oduc s
Spanish Jou nal o Ag icul u al Resea ch Sep embe 2020 • Volume 18 • Issue 3 • e0107
3
Double concen a ion explaining he ou s anding inc ease in Spanish c op p oduc ion
in g ea es demand (Cla e al., 2018), such as ege ables,
oli e oil and wine.
Wi h all o ha , he objec i e o his pape is o analy-
se he changes in Spanish c op p oduc ion du ing he se-
cond hal o he wen ie h cen u y and in es iga e how,
whe e and why i expe ienced such a sus ained and
signi ican g ow h.
Ma e ial and me hods
Da a
In his sec ion, we go in o he main da a sou ces o
analyse Spanish ag icul u al p oduc ion since he 1950s.
Fi s , we collec ed he physical ag icul u al p oduc ion (Qij
exp essed in onnes) om Spain’s Ag a ian S a is ics Yea -
books o 1955, 1980, 2005 and 2010 (MAPAMA, 1955,
1980, 2005, 2010). We wo ked wi h 48 Spanish p o inces
(see Table S1 [suppl.] o he names o he p o inces and
he ela ion o wide adminis a i e egions) and 132 ag i-
cul u al p oduc s. Then we ob ained in o ma ion on he
cu en p ices (Pci exp essed in pese as/ onne) ecei ed
by a me s o each c op (MAPAMA, 1955, 1980, 2005,
2010). No e ha in o de o sepa a e he in la ion impac
and o accoun o he change in eal p oduc ion and in he
alue o c ops, he da a ha e been de la ed, using 1980 as a
base yea . We calcula ed p ice indices o he pe iods 1955-
1980 and 1980-2005/2010 using he 132 c ops included in
he sample and aking 1980 as a base yea . These indices
in o m on he a e age change expe ienced by he p ices o
he c ops du ing bo h ime-spans. Thus, hey can be con-
side ed as a e e ence ha allows e alua ing he changes
in c op p ices o e ime. The calcula ed ag icul u al p ice
indices we e a ound 7.4 and 2.8 o 1955 o 1980 and 1980
o 2005/2010, espec i ely.4 The p ices indices a e used o
ob ain 1955 and 2005/20105 de la ed p ices (Pi exp essed
in cons an 1980 pese as), in oduced below in Eq. (1) o
es ima e mone a y ag icul u al p oduc ion.
Hence, he pape ocuses on h ee key momen s o he
Spanish ag icul u al sec o . Fi s , we look a he mid- wen-
ie h cen u y (1955), jus be o e he 1959 S abiliza ion and
Libe aliza ion Plan men ioned abo e, which b ough an end
o Spanish au a ky and he alded openness o in e na ional
ma ke s. This pe iod also p eceded he apid de elopmen
o hyd aulic in as uc u es du ing he middle yea s o F an-
co’s dic a o ship. Second, as a mid-poin we ocus on 1980,
4 These indices a e e y simila o he ones calcula ed om he ag icul u e’s alue added de la o p o ided by P ados de la Escosu a (2017) ha each 7.6 and 2.5 o 1955 o 1980
and 1980 o 2005/2010, espec i ely. Fo he whole economy, he p ice inc ease was la ge as he indices ob ained om he GDP de la o in P ados (2017) a e 11.4 and 4.6 o 1955
o 1980 and 1980 o 2005/2010, espec i ely.
5 We ake hese yea s o examine changes up o he p esen , since he e was ela i ely high a ia ion in p ecipi a ions in hese yea s, wi h 2005 a ela i ely d y yea and 2010 a
ela i ely humid yea (1955 and 1980 we e bo h a e age yea s).
6 Checking he ep esen a i eness o ou yea s, ag icul u al p oduc ion o 1955 ep esen s 1.02 on 5 yea s a e age ocused on ha yea ; he case o 1980 ep esen s 1.04 on 5 yea s
a e age ocused on ha yea ; 2005 ep esen s 0.94 on 5 yea s a e age ocused on ha yea ; and 2010 ep esen s 1.02 on 5 yea s a e age ocused on ha yea .
7 Fo obse a ions wi h ze o alues, we apply he analy ical limi s a egy when doing he decomposi ion analysis (Ang & Liu, 2007).
when a la ge numbe o wa e in as uc u es had al eady
been de eloped. This momen in ime also wi nessed he
beginning o democ acy in Spain and p epa a ions o he
coun y o join he EEC, which igge ed a g ea comme -
cial expansion and impo an changes o ag icul u e. Fina-
lly, we looked a he beginning o he wen y- i s cen u y by
a e aging a d y (2005) and a humid (2010) yea , ollowing
Cazca o e al. (2015b).6 This inal pe iod pain s a pic u e o
a democ a ic, globally in eg a ed and mo e en i onmen ally
awa e coun y.
Me hodology
A e desc ibing he da a sou ces and cha ac e is ics,
we explain he me hods used o he empi ical analysis.
To s a , we es ima ed mone a y ag icul u al p oduc ion
(P odij exp essed in pese as) o each c op i, p o ince j
and yea . We used in o ma ion on he physical p oduc-
ion (Qij ) and p ices (Pi ) in oduced in he p e ious sec-
ion, which yields:
(1)
F om Eq. (1) we can also ob ain, o each pe iod, he
o al p oduc ion by egion ( ), he o al p o-
duc ion by c op ( ) and he o al na ional p o-
duc ion ( ).
Then, we p oceeded o quan i y he main o ces behind
he his o ical changes in his a iable. This allowed us o
iden i y he mos signi ican s uc u al pa e ns d i ing
ends in he alue o ag icul u al p oduc ion. Wi h his
objec i e, we used decomposi ion analysis, which links
he change o an agg ega e a iable (i.e. mone a y p oduc-
ion in ou case) wi h a se o de e minan s. No e ha he
e ec s de i ed om decomposi ion analysis display he
changes ha would ha e happened i all he o he ac o s
had emained cons an . A p e ious s udy on ag icul u al
pa e ns using decomposi ion analysis o di e en ex-
plana o y ac o s and di e en egional and sec o al de-
ail can be ound in Cazca o e al. (2015b). O he exis-
ing decomposi ion echniques, we u ilized IDA (Ang &
Zhang, 2000); and mo e speci ically, we concen a ed on
he Loga i hmic Mean Di isia Index (LMDI hence o h).
We ollowed Ang (2015), who sugges s using he LMDI-I
addi i e decomposi ion p esen ed in Ang e al. (2010)
when wo king wi h a quan i y indica o .7
𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑖𝑖𝑖𝑖𝑖𝑖 = 𝑄𝑄𝑖𝑖𝑖𝑖𝑖𝑖 𝑃𝑃𝑖𝑖𝑖𝑖
4Miguel Ma ín-Re o illo, Ana Se ano and Ignacio Cazca o
Spanish Jou nal o Ag icul u al Resea ch Sep embe 2020 • Volume 18 • Issue 3 • e0107
No e ha he whole se o e ec s p esen ed in his sec ion
was calcula ed o he wo sub-pe iods o analysis (1955 o
1980 and 1980 o 2005/2010). In o de o ob ain hese im-
pac s o he en i e pe iod (1955 o 2005/2010), we ollowed
Ang e al. (2010), who s a ed ha he chaining p ocedu e
is p e e able when mo e han wo yea s’ wo h o da a a e
a ailable, since i akes be e accoun o he ups and downs
o he pe iod.8 Thus, we go wo se s o decomposi ions
(1950 o 1980 and 1980 o 2005/2010) ha a e chained,
yielding a esul o he whole pe iod (1955 o 2005/2010)
(Ang, 1994; Ang & Liu, 2007; Ang e al., 2010).
Be o e going in o he o mula ion o he decomposi-
ion e ec s, we need o de ine he loga i hmic mean o
mone a y p oduc ion. Using 1 o he inal pe iod and 0
o he ini ial pe iod, we ge :
(2)
A e ha , ollowing Ang e al. (1998) we p oceeded
wi h he decomposi ion o mone a y ag icul u al p oduc-
ion. We i s de eloped he ‘ alue decomposi ion’ and
hen he ‘ ealloca ion decomposi ion’. In bo h, hei mul-
i-dimensional (c op and p o ince dimensions) and mul-
i-le el (di e en le els o agg ega ion) cha ac e allows
an in-dep h examina ion o he long- e m ag icul u al
changes and pa e ns, highligh ing hei main s uc u al
he e ogenei ies.9
Thus, Eq. (3) ep esen s he change in mone a y ag i-
cul u al p oduc ion be ween pe iods 0 and 1.
(3)
This change can also be exp essed as he sum o he
quan i y and p ice e ec s ob ained using he LMDI-I ad-
di i e decomposi ion in oduced by Ang e al. (1998).
This yields Eqs. (4) and (5) o he quan i y (QE) and p i-
ce (PE) e ec s, espec i ely:
(4)
(5)
This i s decomposi ion, he ‘ alue decomposi ion’,
ies o cap u e he ex en o which long- e m changes in
mone a y ag icul u al p oduc ion a e d i en by changes in
he olume o p oduc ion (QE) and by changes in he a-
lue (p ice) o he di e en c ops (PE). No e ha al hough
mone a y ag icul u al p oduc ion has been de la ed o he
whole c op p oduc ion, he p ice e ec exis s as a esul o
he p ice di e ences among di e en p oduc s and migh
8 They li e ally acknowledge ha ‘chaining analysis is p e e ed because i gi es esul s which a e mo e ep esen a i e o he ue si ua ion, i makes ull use o he da a a ailable,
and i is mo e lexible in e ms o applica ion’.
9 Fo a e iew o he impo ance o he mul i-le el and mul i-dimensional cha ac e o IDA, see Ma (2014).
be especially in o ma i e when wo king wi h egional
and c op de ail. This pe spec i e ex ends o he ole o
ag icul u al expansion and o he speci ic c ops and e-
gions esponsible o he inc ease in he alue o Spanish
p oduc ion ha has esul ed om he g owing p oduc ion
o high- alue p oduc s.
Simila ly, in ou second decomposi ion, he change in
mone a y ag icul u al p oduc ion be ween yea s 1 and 0
can be exp essed as he sum o he change in he p o-
duc sha es wi hin p o inces (composi ion e ec , CE),
he change in p o incial composi ion o p oduc ion (sha e
e ec , RE) and he change in o al mone a y p oduc ion
(scale e ec , SE).
=∑∑𝑐𝑐𝑖𝑖𝑖𝑖1∗𝑟𝑟𝑖𝑖1 ∗𝑠𝑠1
𝑖𝑖𝑖𝑖 −∑∑𝑐𝑐𝑖𝑖𝑖𝑖0∗𝑟𝑟𝑖𝑖0 ∗𝑠𝑠0
𝑖𝑖=𝐶𝐶𝐶𝐶+𝑅𝑅𝐶𝐶+𝑆𝑆𝐶𝐶
𝑖𝑖
∆𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃=𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃1−𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃0=∑∑(𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑖𝑖𝑖𝑖1
𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑖𝑖1 ∗𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑖𝑖1
𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃1∗𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃1)−
𝑖𝑖𝑖𝑖 ∑∑(𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑖𝑖𝑖𝑖0
𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑖𝑖0 ∗𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑖𝑖0
𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃0∗𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃0)
𝑖𝑖𝑖𝑖
∆𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃=𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃1−𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃0=∑∑(𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑖𝑖𝑖𝑖1
𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑖𝑖1 ∗𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑖𝑖1
𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃1∗𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃1)−
𝑖𝑖𝑖𝑖 ∑∑(𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑖𝑖𝑖𝑖0
𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑖𝑖0 ∗𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑖𝑖0
𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃0∗𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃0)
𝑖𝑖𝑖𝑖
=∑∑𝑐𝑐𝑖𝑖𝑖𝑖1∗𝑟𝑟𝑖𝑖1 ∗𝑠𝑠1
𝑖𝑖𝑖𝑖 −∑∑𝑐𝑐𝑖𝑖𝑖𝑖0∗𝑟𝑟𝑖𝑖0 ∗𝑠𝑠0
𝑖𝑖=𝐶𝐶𝐶𝐶+𝑅𝑅𝐶𝐶+𝑆𝑆𝐶𝐶
𝑖𝑖
(6)
Again, ollowing Ang e al. (1998), we es ima ed he
CE, RE and SE in he ollowing equa ions:
𝐶𝐶𝐶𝐶=∑∑𝐿𝐿(𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃
𝑖𝑖𝑖𝑖1
,𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃
𝑖𝑖𝑖𝑖0
)
𝑖𝑖ln(𝑐𝑐𝑖𝑖𝑖𝑖1
𝑐𝑐𝑖𝑖𝑖𝑖0)⁡
𝑖𝑖
(7)
𝑅𝑅𝑅𝑅=∑∑𝐿𝐿(𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑖𝑖𝑖𝑖1,𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑖𝑖𝑖𝑖0)
𝑖𝑖ln(𝑃𝑃𝑖𝑖1
𝑃𝑃𝑖𝑖0)⁡
𝑖𝑖
(8)
𝑆𝑆𝑆𝑆 =∑∑𝐿𝐿(𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑖𝑖𝑖𝑖1,𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑖𝑖𝑖𝑖0)
𝑖𝑖ln(𝑠𝑠1
𝑠𝑠0)⁡
𝑖𝑖
(9)
Acco dingly, we can explain he changes in mone a y
p oduc ion as a esul o a ia ions in: 1) he c op com-
posi ion o p oduc ion (CE) ha allows us o e alua e he
p oduc i e change owa d highe - o lowe - alue p oduc s
wi hin egions; 2) he egional edis ibu ion o p oduc-
ion (RE) ha p o ides in o ma ion on he geog aphical
displacemen o mone a y p oduc ion be ween egions;
and 3) he scale o mone a y p oduc ion (SE). Hence, he
‘ ealloca ion decomposi ion’ sheds ligh on he con ibu-
ion o s uc u al changes, i.e. on he changes in c op and
egional specializa ion pa e ns.
Resul s
The Spanish ag icul u al p oduc ion: a iew
based on egions and p oduc s
Table 1 shows he le el o p oduc ion in mone a y uni s
and i s annual g ow h a es. We can obse e con inuous
g ow h in Spanish ag icul u al p oduc ion o almos 2%
𝐿𝐿(𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑖𝑖𝑖𝑖1,𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑖𝑖𝑖𝑖0)=𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑖𝑖𝑖𝑖1− 𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑖𝑖𝑖𝑖0
ln(𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑖𝑖𝑖𝑖1)−ln⁡(𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑖𝑖𝑖𝑖0)
∆𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃 = 𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃1−𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃0= ∑∑𝑄𝑄𝑖𝑖𝑖𝑖1𝑃𝑃𝑖𝑖1
𝑖𝑖𝑖𝑖 −∑∑𝑄𝑄𝑖𝑖𝑖𝑖0𝑃𝑃𝑖𝑖0
𝑖𝑖=𝑄𝑄𝑄𝑄+𝑃𝑃𝑄𝑄
𝑖𝑖
𝑄𝑄𝑄𝑄=∑∑𝐿𝐿(𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑖𝑖𝑖𝑖1,𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑖𝑖𝑖𝑖0)
𝑖𝑖ln(𝑄𝑄𝑖𝑖𝑖𝑖1
𝑄𝑄𝑖𝑖𝑖𝑖0)⁡
𝑖𝑖
𝑃𝑃𝑃𝑃 =∑∑𝐿𝐿(𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑖𝑖𝑖𝑖1,𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑖𝑖𝑖𝑖0)
𝑖𝑖ln(𝑃𝑃𝑖𝑖1
𝑃𝑃𝑖𝑖0)⁡
𝑖𝑖
Spanish Jou nal o Ag icul u al Resea ch Sep embe 2020 • Volume 18 • Issue 3 • e0107
5
Double concen a ion explaining he ou s anding inc ease in Spanish c op p oduc ion
annually o e mo e han 50 yea s. The highes g ow h was
om 1955 o 1980, when he igu e inc eased by 3.3% an-
nually. This g ow h coincides wi h he widesp ead adop-
ion o chemical p oduc s and ag icul u al machine y and
he cons uc ion o majo hyd aulic in as uc u es du ing
he F anco dic a o ship.10 Howe e , g ow h was slowe
om 1980 o 2005/2010 (0.7% annually). E en hen,
howe e , Spanish ag icul u e inc eased i s p oduc ion, in
con as o o he Eu opean coun ies, whe e p oduc ion
emained s able om he mid-1980s.
Table 1 and Fig. 1 show ha c op p oduc ion was
concen a ed in he sou h and eas o Spain, especially in
Andalusia and Mu cia; meanwhile, he no hwes los im-
po ance in e ms o ag icul u al p oduc ion. I a he be-
ginning o he pe iod he egion o Andalusia had a sha e
o 20.2% o Spanish c op p oduc ion, and 23.5% in 1980,
by he end o he pe iod his had isen o 31.9%. O he
egions –like A agon, Ex emadu a and Mu cia– subs-
an ially inc eased hei own impo ance in Spanish c op
p oduc ion. This concen a ion coincides wi h a pe iod o
as cons uc ion o hyd aulic in as uc u es du ing he
F anco dic a o ship, pa icula ly in he Medi e anean
a ea and he Eb o basin. By con as , he wo egions o
10 The e was also some cons uc ion o hyd aulic in as uc u es be o e he F anco dic a o ship, especially om 1911 o 1936. Howe e , he mos signi ican building was in he
1950s, 1960s and 1970s (He anz, 2004).
Cas ile and Leon and Galicia, bo h si ua ed in he no -
hwes , los almos hal o hei weigh ing in o al c op
p oduc ion.
Looking a Fig. 1, he la ges inc eases in Spanish c op
p oduc ion occu ed in he p o inces o Alme ia, Badajoz,
Mu cia, Se ille, Za agoza, Huesca, Valencia, Co doba,
G anada and Le ida, being many o hem in he eas , and
se e al in he sou h o Spain. The case o Alme ia – he
mos sou h-eas e ly p o ince and he mos a id – is s i-
king. The ole o inc eased i iga ion in his wa e -sca ce
egion (albei a egion wi h good soil and plen y o suns-
hine hou s) is a dominan explana o y ac o in he ag i-
cul u al e olu ion ha occu ed. He e we ind a yea ly
g ow h in c op p oduc ion o 6.3% h oughou he pe iod.
The di e en specializa ion o each p o ince p odu-
ced hese di e en ial g ow hs. In o de o ob ain a com-
ple e iew, i is necessa y also o obse e he e olu ion
o p oduc ion in e ms o g oups o p oduc s; his helps
us o unde s and he changing c op pa e ns o Spanish
ag icul u e. As Table 2 shows, he bigges g ow h is o
be ound in ege ables, which mo ed om making up
10% o he o al p oduc ion alue in 1955 o 32% in
2005/2010. Table 2 also shows inc eases in he sha e o
Table 1. Mone a y p oduc ion alue (p ices de la ed using he base yea 1980) by egion (in million pese as)
and annual g ow h a es
Sou ces: Own elabo a ion om MAPAMA (1955, 1980, 2005, 2010).
Le el (million pese as) Annual g ow h a es (%)
Region 1955 1980 2005/2010 1955 o
1980 1980 o
2005/2010 1955 o
2005/2010
AN Andalusia 84,854 224,431 363,687 4.0% 1.8% 2.8%
AR A agon 22,814 64,116 93,543 4.2% 1.4% 2.8%
AS As u ias 8,307 9,472 6,541 0.5% -1.4% -0.5%
CB Can ab ia 2,141 2,887 1,880 1.2% -1.6% -0.2%
CL Cas ile and Leon 79,737 132,993 114,888 2.1% -0.5% 0.7%
CM Cas ile-La Mancha 44,508 119,454 107,900 4.0% -0.4% 1.7%
CT Ca alonia 31,577 87,691 77,362 4.2% -0.5% 1.7%
EX Ex emadu a 22,137 53,936 93,546 3.6% 2.1% 2.8%
GA Galicia 34,348 56,970 54,056 2.0% -0.2% 0.9%
IB Balea ic Islands 5,929 13,097 10,791 3.2% -0.7% 1.2%
MC Mu cia 7,836 32,340 72,200 5.8% 3.0% 4.4%
MD Mad id 6,592 9,640 5,783 1.5% -1.9% -0.3%
NC Na a a 10,855 22,694 26,713 3.0% 0.6% 1.7%
PV Basque Coun y 6,917 8,722 7,163 0.9% -0.7% 0.1%
RI La Rioja 6,821 16,143 15,035 3.5% -0.3% 1.5%
VC Valencian Comm. 44,785 101,584 89,958 3.3% -0.4% 1.4%
TOT To al 420,158 956,170 1,141,047 3.3% 0.7% 1.9%

6Miguel Ma ín-Re o illo, Ana Se ano and Ignacio Cazca o
Spanish Jou nal o Ag icul u al Resea ch Sep embe 2020 • Volume 18 • Issue 3 • e0107
oli e oil and wine ( he la e up un il 1980). Bo h ci us
and non-ci us ui s mo ed om a sha e o abou 4% o
abou 7–8% a he end o he pe iod11 . All o hese p o-
duc s a e high alue-added c ops. Besides, odde c ops
also inc eased in impo ance in c op p oduc ion, mo ing
om a sha e o 6% o 11%.
We ha e hen clea ly seen his double concen a ion:
ealloca ion o he sou h and eas o Spain, and specializa-
ion in high alue-added p oduc s. This came abou hanks
o se e al ac o s. Fi s , Spanish expo s o ege ables and
ui ose d ama ically o become he leading ag icul u al
expo p oduc s in ecen yea s; hey a e highly p o i able
and a e he clea es ep esen a ion o i iga ed p oduc ion
(pa icula ly in he sou heas o Spain). This is no a new
specializa ion pa e n: Spain has adi ionally expo ed
ui and ege ables e e since he las decades o he ni-
ne een h cen u y, when i s expo s ep esen ed app oxi-
ma ely one hi d o he global expo s o Medi e anean
ho icul u al p oduc s. Du ing he second globaliza ion,
he di e si ica ion o ag icul u al expo s and he g ea e
impo ance o ans o med p oduc s in in e na ional ade
did no change he p eponde ance ha ui and ege a-
bles ha e enjoyed in he ag icul u al sec o . The ela i-
e and absolu e inc ease in he p oduc ion o ege ables,
ui and oil was also d i en by in e nal demand. The i-
sing Spanish popula ion necessi a ed highe p oduc ion
o mee he ood needs. Besides, economic de elopmen
11 In he case o ci us ees, Valencia is one o he main p oduce s o hese p oduc s, despi e small size o hese a ms (Reig & Picazo, 2004).
igge ed imp o emen s in he Spanish die , which mo-
ed owa ds g ea e consump ion o ui , ege ables and
oil. In he case o wine, in e nal Spanish demand ell
ab up ly in he second pe iod, despi e i s inc easing ex-
po o ien a ion. Thus, Spanish wine p oduc ion lowed
inc easingly o he in e na ional ma ke s. These dec eases
in p oduc ion in a ew p oduc s go o explain he igu e
in Table 2 o o al p oduc ion in he las pe iod, which
did inc ease compa ed o he 1980s, bu a a mo e modes
a e han be o e.
The changing die o Spania ds also in ol ed g ea e
mea consump ion. Mo eo e , he e was a big inc ease
in mea expo s, especially a e he 1980s. This was e-
lec ed in he g ea e impo ance o li es ock p oduc ion
du ing he second globaliza ion, which a he same ime
equi ed la ge olumes o animal eed, bo h domes ica-
lly p oduced and impo ed. Du ing he ea ly decades o
he second hal o he wen ie h cen u y, he e was also a
ma ked inc ease in he p oduc ion o non-whea ce eals
o animal eed, such as ba ley and maize. As illus a ion,
a he beginning o he 1960s, 39% o ce eal p oduc ion
was o animal eed, bu by he beginning o he 1980s
his igu e had eached 72%. Ne e heless, as Table 2
shows, o e he pe iod as a whole he e was a big decli-
ne in he sha e o ce eals – om 47% in 1955 o 14% in
2005/2010. The decline in some low- alue p oduc s (such
as legumes, and o he woody c ops) was also s iking; his
Figu e 1. Mone a y alue o p oduc ion in 1955 (size o he ci cles), absolu e changes
be ween 1955 and 2005/2010 (colou in ensi y o he ci cles) and ela i e (%) change be ween
1955 and 2005/2010 (colou o he p o inces). Table S1 [suppl.] o he names o he p o in-
ces and he ela ion o wide adminis a i e egions. Sou ce: Own elabo a ion om MAPAMA
(1955, 1980, 2005, 2010).
Spanish Jou nal o Ag icul u al Resea ch Sep embe 2020 • Volume 18 • Issue 3 • e0107
7
Double concen a ion explaining he ou s anding inc ease in Spanish c op p oduc ion
was mainly due o a all in hei consump ion, especially
a e 1980.
Causes o he double concen a ion: decomposi ion
esul s
In his sec ion, we quan i y and expand on he causes
o he double concen a ion seen in p e ious sec ions.
To ha end, i s we examined he alue decomposi ion,
which analyses changes in he olumes and p ices as
d i e s o mone a y p oduc ion. Then we looked a he
de e minan s ha igge ed he inc eases in c op p oduc-
ion. As indica ed in he me hodology, he eason o he
exis ence o a p ice e ec is ha , al hough he e ec o
p ices is emo ed a he agg ega e le el o whole c op
p oduc ion, he indi idual p ices o p oduc s e ol ed in
di e en ways. These di e ences may show hose c ops
ha expe ienced highe inc eases in hei alue added,
which hus encou aged hei p oduc ion and gene a ed
addi ional income.
In Fig. 2a we ound a nega i e p ice e ec o ce eals
om 1955 o 1980, gi en ha ce eal p ice ell o emai-
ned s able. Howe e , we obse e a posi i e quan i y e ec
12 Ne e heless, he a idi y o hese p o inces gene a es p oblems o a ailabili y o esou ces, especially wa e (Cazca o e al., 2015b).
(i.e. he e was an inc ease in physical uni s o p oduc ion),
which con ibu ed o a ne inc ease in he alue o p o-
duc ion. In egional e ms, his happened in Cas ile-La
Mancha and no ably Cas ile and Leon, wo a eas wi h sig-
ni ican ce eal p oduc ion.
Despi e he gene ally s onge impac o he quan i y
e ec , he p ice e ec explained mo e han 33% o he
change, pa icula ly o oli e oil and ege ables. Anda-
lusia accoun s o an eno mous sha e o hese g oups o
p oduc s. In he case o oli e oil, he e olu ion o i s p o-
duc ion and expo was sligh ly e a ic be ween 1980 and
he i s yea s o he wen y- i s cen u y.
As o ege ables, du ing he las decades o ou pe-
iod o s udy, hei p oduc ion kep inc easing ia quan i y
e ec s, and p ice e ec s showed ela i ely g ea e impo -
ance. These changes we e geog aphically concen a ed
in Andalusia, Ex emadu a and Mu cia (Ex emadu a is
mo e in he wes , bu he e was an impo an concen a-
ion in he sou h and eas o Spain), a eas wi h op imal
condi ions o hese c ops: a high numbe o sunshine
hou s pe yea and wa e esou ces om he expansion
o i iga ion.12 Fu he mo e, inc eased ex e nal demand
o his p oduc ype gene a ed g owing expo s. The po-
si i e p ice e ec was a s ong incen i e o p oduce mo e
Table 2. Mone a y p oduc ion alue (p ices de la ed using he base yea 1980) by p oduc g oup (in
million pese as) and annual g ow h a es (%)
[1] Oil e e s o oli e oil, oli e consump ion a e oli es ha a e di ec ly consumed, wine is conside ed as
an indi idual i em. G apes and aisins o consump ion a e included as non-ci us ui s. Sou ce: Own
elabo a ion om Spain’s Ag a ian S a is ics Yea books o 1955, 1980 and 2005 o 2010 (MAPAMA,
1955, 1980, 2005, 2010).
P oduc g oup[1]
Mone a y p oduc ion alue Annual g ow h a e (%)
1955 1980 2005/2010 1955-
1980 1980-
2005/10 1955-
2005/10
Oil 22,143 55,483 127,276 3.74 3.12 3.42
Ce eals 195,687 270,878 156,887 1.31 -2.00 -0.42
Indus ial c ops 25,422 69,295 33,185 4.09 -2.69 0.51
Fodde 26,447 85,907 124,578 4.83 1.39 3.03
Ci us ui ees 16,011 46,039 84,135 4.32 2.26 3.24
Non-ci us ui ees 16,602 73,739 95,732 6.15 0.97 3.43
Nu s 10,276 23,964 42,915 3.44 2.18 2.79
Vege ables 43,739 170,987 363,843 5.60 2.84 4.16
Legumes 21,836 16,488 7,538 -1.12 -2.86 -2.02
Oli e consump ion 1,954 5,372 13,851 4.13 3.57 3.84
O he woody c ops 4,444 2,418 1,114 -2.41 -2.83 -2.63
Tube s 32,878 58,854 30,504 2.36 -2.40 -0.14
Wine 2,719 76,747 59,489 14.29 -0.94 6.11
To al 420,158 956,170 1,141,047 3.34 0.66 1.94
8Miguel Ma ín-Re o illo, Ana Se ano and Ignacio Cazca o
Spanish Jou nal o Ag icul u al Resea ch Sep embe 2020 • Volume 18 • Issue 3 • e0107
ege ables, subs i u ing low alue-added o high a-
lue-added p oduc s, which e eals concen a ion in e ms
o p oduc s.
The case o odde c ops is also s iking. Despi e a
all in he quan i y o p oduc ion, p ices boos ed he a-
lue o p oduc ion. This happened in no he n egions,
such as As u ias, Can ab ia, Vizcaya, Guipúzcoa and
o he no he n p o inces (see Fig. S1a [suppl.]) wi h
high ain all and high yields. Finally, he alue o wine
p oduc ion ell since he 1980s – because o bo h he
quan i y and p ice e ec s, al hough he quan i y e ec
was posi i e om 1955 o 1980. This dec ease in mo-
ne a y wine p oduc ion om 1980 (Table 2 and Fig. 2a)
hinde s la ge egional he e ogenei ies. Acco ding o he
in o ma ion om he Spain’s Ag a ian S a is ics Yea -
books o 1980 and 2005 o 2010 (MAPAMA, 1955,
1980, 2005 & 2010) wine p oduc ion no ably inc eased
in he no h/no heas o he coun y ( egions o La Rio-
ja, Na a a, Ba celona and Huesca). Howe e , be ween
1980 and 2005/2010 impo an dec eases happened in
Cas illa-La Mancha ( o example in he p o ince o Ciu-
dad Real), d i ing he esul s a he na ional le el. Ca-
ñiza es Ruiz & Ruiz Pulpón (2014) explained ha he
13 To e alua e he obus ness o hese esul s we calcula ed he alue o c op p oduc ion in 1955, 1980 and 2005/2010 a cons an 1980 p ices and applied he same decomposi ion
as in Table 3. The decomposi ion o he alue o c op p oduc ion using de la ed p ices (Table 3) is e y simila o ha using 1980 cons an p ices. Conc e ely, o he pe iod 1955-
2005/2010 he di e ences be ween bo h app oaches ange a ound 1% and 2,5% o he di e en e ec s. Thus, he decomposi ion is obus o he selec ion o a cons an o de la ed
base yea .
14 This numbe ob iously ma ches he di e ences shown in Tables 1 and 2 o alue in 2005/2010 o 1,141,047 million pese as and in 1955 o 420,158 million pese as.
EC egula ion en ailed he loss o app ox. 130,000 ha o
ineya ds in Cas illa-La Mancha om 1987 o 1999 (a
all o a ound 20% o he exis ing a ea). Despi e ha a
he same ime he e was a conside able expansion o i i-
ga ed su ace, i was no enough o o se he huge all o
ain- ed a ea.
The o me esul s can be complemen ed wi h he
quan i ica ion o he changes in egional and c op compo-
si ion as de e minan s o he change in Spanish c op p o-
duc ion. In Table 313 we obse e ha he Spanish alue
o c op p oduc ion inc eased om 1955 o 2005/2010.14
The indings om LMDI decomposi ion indica e ha his
g ow h was essen ially (84%) explained by he SE. The
CE (16%) also goes some way o explaining he chan-
ge in he alue o p oduc ion; while he sha e e ec was
almos non-exis en ac oss his long pe iod. Di e en-
ia ing be ween he i s and he second pe iod s udied,
we obse e how he scale e ec was la ge in he se-
cond pe iod, while he composi ion e ec was la ge in
he i s . Al hough he igu es a e small, i is in e es ing
ha egional changes led o an inc ease in he alue o
mone a y p oduc ion in he i s pe iod, bu o a dec ease
in he second.
Figu e 2. Decomposi ion o he changes (in million pese as) in o he p ice and quan i y e ec s (a) and in o he composi ion, sha e
and scale e ec s by egion (b). AN: Andalusia; AR: A agon; CL: Cas ile and Leon; CM: Cas ile-La Mancha; CT: Ca alonia; EX:
Ex emadu a; GA: Galicia; MC: Mu cia; and VC: Valencian Communi y. ‘O he ’ comp ises small egions (AS: As u ias; CB: Can-
ab ia; IB: Balea ic Islands; MD: Mad id; NC: Na a a; PV: Basque Coun y; and RI: La Rioja) and hei g aph is shown in Table S1
[suppl.]. Sou ce: Own elabo a ion (MAPAMA, 1955, 1980, 2005, 2010).
Spanish Jou nal o Ag icul u al Resea ch Sep embe 2020 • Volume 18 • Issue 3 • e0107
9
Double concen a ion explaining he ou s anding inc ease in Spanish c op p oduc ion
We saw in Fig. 2a ha mone a y p oduc ion inc ea-
sed in mos egions; in Fig. 2b, we see ha his is la gely
explained by scale e ec s, bu ha he posi i e composi-
ion e ec s also play a pa . P oduc ion ended o mo e o
mo e alued p oduc s o high alue-added p oduc s. As in
Fig. 2a, he case o Andalusia is e y illus a i e. This e-
gion led he g ow h in alue. We can see ha al hough o
his egion he e a e pu ely scale e ec s (mos egions in-
c ease hei mone a y p oduc ion), he e a e also no able
egional/sha e e ec s (p oduc ion mo ed o his egion)
and p oduc composi ion e ec s (highe specializa ion in
mo e alued p oduc s).
In his ega d, he spa ial concen a ion had a nega i-
e sha e e ec in e ms o p oduc ion o egions such as
Cas ile and Leon, Cas ile-La Mancha, Valencian Commu-
ni y, Can ab ia, and many o he s such as Galicia, Mad id,
he Basque coun y,15 e c. This means ha hese egions
educed he p oduc ion o some c ops, no because ho-
se c ops we e less in e idence in Spain’s p oduc ion as
a whole, bu because p oduc ion mo ed elsewhe e. As
hin ed abo e, his ‘elsewhe e’ included Andalusia, Ex-
emadu a, Mu cia and A agon. Al hough globally sha e
e ec s seem o ha e had a small in luence on changes in
p oduc ion, hey ha e impo an explana o y powe when
we look a he egional de ail.
Discussion
In his wo k, we ha e analysed he s uc u al changes
in ag icul u al p oduc ion in Spain du ing he second hal
o he wen ie h cen u y and he beginning o he wen-
y- i s , in he amewo k o majo ans o ma ions in he
ag icul u al sec o i sel , as well as in he ins i u ional and
poli ical con ex . We assessed how changes in he olu-
me and p ice o p oduc ion ha e condi ioned he alue
o c op p oduc ion in he long e m. Then, we examined
changes in c op specializa ion and egional composi ion
as d i e s o mone a y c op p oduc ion. We ound s ong
e idence o he exis ence o a ‘double concen a ion’, in
space and c ops.
15 See o he small egions (AS: As u ias; CB: Can ab ia; IB: Balea ic Islands; MD: Mad id; NC: Na a a; PV: Basque Coun y; and RI: La Rioja) Fig. S1b [suppl.]
16 Se e al di e ences a e ound be ween he ype o he i iga ion (Rod íguez-Díaz e al., 2008).
The highly de ailed egional and p oduc disagg e-
ga ion is an impo an alue added o his pape , as i
allows in-dep h s udy o he long- e m changes expe-
ienced in he Spanish ag icul u al sec o . In e ms o
me hodology, a e comple ing he da abase and de-
i ing he main desc ip i e ajec o ies o ag icul u-
al p oduc ion, we used index decomposi ion analysis
(IDA). This echnique is use ul in p o iding insigh s
in o he ac o s ha lie behind he s uc u al changes
expe ienced by he Spanish ag icul u al sec o . Ou
a icle includes empi ical e idence on he ac o s ha
enabled Spanish ag icul u al p oduc ion o g ow so
conspicuously in Eu opean e ms. We obse e ha he
he e ogeneous g ow h in Spanish c op p oduc ion was
based on a double concen a ion: ag icul u e ocused on
high alue-added p oduc s, such as ege ables, ui and
oli e oil; and, spa ially i was concen a ed in he sou h
and eas o Spain. The expansion o i iga ion, he e o-
lu ion o in e nal demand and he p og essi e in eg a ion
in o in e na ional ma ke s all go o explain his double
concen a ion.
In pa icula , c op p oduc ion inc eased a 2% yea ly
om 1955 o 2005/2010; his g ow h was pa icula ly in-
ensi e be ween 1955 and 1980, a pe iod when he an-
nual g ow h a e s ood a 3.3%. Thus, p oduc ion soa ed
(as e ealed by he scale e ec ) and en ailed a spa ial
concen a ion, no ably in he sou h and eas o Spain (as
shown by he sha e e ec ). Andalusia showed i s g owing
impo ance in ag icul u al p oduc ion, wi h a sha e ha
wen om 20.2% o he alue o p oduc ion in 1955 o
31.9% a he beginning o he wen y- i s cen u y. The
ela i e inc ease mos ly occu ed om 1980. This egion
bols e ed he adi ion o cul i a ing p oduc s o expo
and p oduc s wi h high alue-added (Se ano & Pinilla,
2014; He nández e al., 2016; Apa icio e al., 2018). The
expansion o i iga ion acili a ed inc eased ege able
p oduc ion (6% annually h oughou he pe iod s udied)
and oli e p oduc ion (3.6% annually)16 . O he egions
like Ex emadu a, Mu cia o A agon also specialized in
p oduc s ha enjoyed s ong demand, acqui ing a g ea e
sha e o c op p oduc ion.
Table 3. Loga i hmic Mean Di isia Index (LMDI-I) mul iplica i e decomposi ion o alue o c op p oduc ion
changes (million pese as)
CE: composi ion e ec . RE: sha e e ec . SE: scale e ec . Sou ce: Own elabo a ion om MAPAMA (1955, 1980,
2005, 2010).
CE (%) RE (%) SE (%) P od
1955 o 1980 87,016 16 861 0.2 448,136 84 536,012
1980 o 2005/2010 25,136 14 -1064 -0.6 160,804 87 184,877
1955 o 2005/2010 112,152 16 -203 0.0 608,940 84 720,889